With the acceleration of urbanization, the integration of renewable energy and advanced energy storage technologies holds great promise for improving energy efficiency in urban rail transit systems. However, this also brings challenges in energy system planning, configuration, and operation optimization. By analyzing the energy demand of traction loads and the dynamic characteristics of intermittent photovoltaic output within the rail transit system, a robust model for energy storage configuration and optimized operation under the polyhedron uncertainty set of source load was established. The objective function, subject to stability and economic constraints, aims to minimize the comprehensive daily average cost of energy storage. A column-and-constraint generation algorithm is used to solve the model. The simulation example assesses the impact of uncertainty on the scheduling operation plan and the daily total comprehensive cost, thereby validating the effectiveness of the proposed optimization model. The study has significant implications for understanding the role of energy storage systems as a key component in improving the resilience and stability of urban rail transit networks.